Application of in silico and TTC based decision tree for safety assessment of botanicals
Safety assessment of botanicals has always been a challenge, not only due to the difficulties in defining them analytically, but also due to data gaps in traditional toxicology endpoints. A recently proposed decision tree by Little et al. nicely integrates the 21st century toxicology concepts to address the data gaps and identify next steps, apart from emphasising the need for well characterisation of the substance before proceeding with their safety assessments.
In the past decade botanical/herbal ingredients are being increasingly used in various consumer preparations such as cosmetics and food, making their safety assessments critical. However, traditional safety/risk assessment approach has often been challenging due to lack of well characterised analytical data, gaps for endpoint of concern along with practical challenges of conducting their toxicological studies, costs and animal usage. In this regard, we recently came across an interesting article of Little et al, which describes a structured approach to overcome many of these obstacles by applying state-of-the-art analytical techniques followed by application of a decision tree using food intake levels and established in silico toxicology assessment tools to identify hazards for the individual chemical constituents.
As per the proposed approach, each chemical constituent is: (a) quantitatively benchmarked against similar constituents in commonly consumed foods or botanicals with well-established safety profiles, (b) systematically evaluated for toxicity data utilizing structure-activity relationships, and, (c) compared to established thresholds of toxicological concern in absence of safety data or structural analogues. Finally, where safety endpoint gaps still exist and cannot be resolved following the decision tree, then the identification of potential constituents of concern together with the well-characterised botanical will help in deciding the appropriate study design or testing strategy.
The application of the decision tree has been illustrated in the article with help of three scenarios: (1) by benchmarking exposures to constituents commonly found for example in food to exposures via supplement use (e.g., isoflavone exposure via supplements vs soy and other foods); (2) by using constituent analysis to justify bridging safety data between different methods of botanical preparation; (3) and by establishing exposure thresholds for individual constituents with limited human use data using in silico toxicology assessment tools or TTC approach.
To conclude, although the concepts applied in the proposed decision tree are not new and are in use in other sectors, the decision tree surely provides a systematic methodology to guide the safety assessor through the many variables that require consideration. Another critical aspect which has not been discussed in detail in the article, but remains a challenge in general while performing the safety assessments of botanical ingredients, is to address the possibility for influences of the matrix effects. Therefore, it would be important to evaluate this aspect as part of every botanical safety evaluation, by identifying the ingredients with similar chemical structures/congeners, in order to anticipate similar mechanism of actions. Similarly, one has to also take into account dose additions for constituents with similar structures, while performing TTC based assessments.
